Now showing 1 - 2 of 2
  • Publication
    PARNN: A Probabilistic Autoregressive Neural Network Framework for Accurate Forecasting
    (2022) ; ;
    Panja, Madhurima
    Kumar, Uttam
    Forecasting time series data represents an emerging field of research in data science and knowledge discovery with vast applications ranging from stock price and energy demand prediction to the early prediction of epidemics. Numerous statistical and machine learning methods have been proposed in the last five decades with the demand for high-quality and reliable forecasts. However, in real-life prediction problems, situations exist in which a model based on one of the above paradigms is preferable. Therefore, hybrid solutions are needed to bridge the gap between classical forecasting methods and modern neural network models. In this context, we introduce a Probabilistic AutoRegressive Neural Network (PARNN) model that can handle a wide variety of complex time series data (e.g., nonlinearity, non-seasonal, long-range dependence, and non-stationarity). The proposed PARNN model is built by creating a fusion of an integrated moving average and autoregressive neural network to preserve the explainability, scalability, and "white-boxlike" prediction behavior of the individuals. Sufficient conditions for asymptotic stationarity and geometric ergodicity are obtained by considering the asymptotic behavior of the associated Markov chain. Unlike advanced deep learning tools, the uncertainty quantification of the PARNN model based on prediction intervals is obtained. During computational experiments, PARNN outperforms standard statistical, machine learning, and deep learning models (e.g., Transformers, NBeats, DeepAR, etc.) on a diverse collection of real-world datasets from macroeconomics, tourism, energy, epidemiology, and others for short-term, medium-term, and long-term forecasting. Multiple comparisons with the best method are carried out to showcase the superiority of the proposal in comparison with the state-ofthe-art forecasters over different forecast horizons.
      55  2
  • Publication
    Stochastic forecasting of COVID-19 daily new cases across countries with a novel hybrid time series model
    (2022) ;
    Rai, Shesh N.
    Bhattacharyya, Arinjita
    An unprecedented outbreak of the novel coronavirus (COVID-19) in the form of peculiar pneumonia has spread globally since its first case in Wuhan province, China, in December 2019. Soon after, the infected cases and mortality increased rapidly. The future of the pandemic’s progress was uncertain, and thus, predicting it became crucial for public health researchers. These predictions help the effective allocation of health-care resources, stockpiling, and help in strategic planning for clinicians, government authorities, and public health policymakers after understanding the extent of the effect. The main objective of this paper is to develop a hybrid forecasting model that can generate real-time out-of-sample forecasts of COVID-19 outbreaks for five profoundly affected countries, namely the USA, Brazil, India, the UK, and Canada. A novel hybrid approach based on the Theta method and autoregressive neural network (ARNN) model, named Theta-ARNN (TARNN) model, is developed. Daily new cases of COVID-19 are nonlinear, non-stationary, and volatile; thus, a single specific model cannot be ideal for future prediction of the pandemic. However, the newly introduced hybrid forecasting model with an acceptable prediction error rate can help healthcare and government for effective planning and resource allocation. The proposed method outperforms traditional univariate and hybrid forecasting models for the test datasets on an average.
    Scopus© Citations 2  86  4